No abstract

Cyclophosphamide (CY)-induced neutropenia exacerbates septic shock and acute lung injury during Candida albicans (CA) fungemia in conscious rats. We hypothesized that treatment of such animals with recombinant murine granulocyte-macrophage colony-stimulating factor (GM-CSF) improves host defense during disseminated candidiasis by increasing peripheral neutrophils (PMNs) and enhancing endogenous production of antifungal cytokines including tumor necrosis factor-alpha (TNF). Naive (neutrophil-replete) or neutropenic rats were infected with 10(7) yeast-phase CA; subgroups received GM-CSF (25 micrograms/kg sc) or sterile 0.9% NaCl (NS) twice a day beginning 3 days before CA infection. Arterial hemodynamics, formed blood elements, bioactive TNF in serum and bronchoalveolar lavage fluid (BALF), and lung histopathology were monitored for up to 72 h after infection. All naive animals receiving GM-CSF (n = 5) and 78% of naive rats given NS (n = 9) remained normotensive through 72 h with no lung injury, differing principally in baseline PMNs before CA infection (8.8 +/- 1.8 x 10(3)/microliters, mean +/- SE, vs. 3.7 +/- 0.4 x 10(3)/microliters, respectively, P < 0.01). Neutropenic rats given NS (baseline PMN = 41 +/- 10/microliters, n = 7) were sensitized to CA, and 100% died of hypothermic shock with severe respiratory distress within 56 h of infection. Pulmonary periarterial and alveolar hemorrhage were prominent. Although GM-CSF did not increase baseline PMNs in CY animals by the outset of infection (162 +/- 58/microliters, n = 8), 62% of these rats remained normotensive and eupneic through 72 h (P < 0.01), and their lungs showed no perivascular hemorrhage, alveolar disruption, or fungi.(ABSTRACT TRUNCATED AT 250 WORDS)

Candida albicans (CA) increasingly causes septic shock, acute lung injury, and multiple organ damage during immunosuppression-related neutropenia. However, the effects of neutrophil (PMN) depletion on induction of tumor necrosis factor-alpha (TNF) by CA and its potential mediation of Candida septic shock are unknown. We hypothesized that reduced CA uptake by circulating PMNs during cyclophosphamide (CY)-related neutropenia sensitizes to TNF-mediated shock from enhanced cytokine production after phagocytosis by tissue macrophages. Absolute or relative neutropenia (PMNs < or = 500/microliters or 2,500/microliters) was modeled in rats by intraperitoneal CY 4-8 days before 10(9) yeast-phase CA (acute studies < or = 24 h, n = 81 animals) or 10(6) CA (subacute studies < or = 72 h, n = 25). Compared with neutrophil-sufficient rats, absolute neutropenia accelerated hemodynamic collapse and respiratory distress after 10(9) CA, and pulmonary microvascular permeability was amplified. These changes evolved without increased candidemia or elevations in bioactive or antigenic serum TNF, which remained low even at death (42.3 +/- 14.8 U/ml vs. 12.6 +/- 2.9 U/ml for CY + saline, means +/- SE, P = NS). In contrast, significant TNF in lung tissue and bronchoalveolar lavage fluid (BALF) was evident within 6 h in CY + 10(9) CA rats. Electron microscopy confirmed hyphal proliferation into alveoli from yeast within mononuclear cells in lung capillaries. Subacute disseminated candidiasis after 10(6) CA was not associated with elevated serum, lung, or BALF TNF. We conclude that differential systemic and intrapulmonary TNF production occur in CA septic shock during preexisting neutropenia, with compartmentalized TNF production in the lower respiratory tract accompanying yeast-mycelial transformation. Thus TNF is not an obligate mediator of acute candidemic shock or subacute disseminated candidiasis during CY-induced immunosuppression but may initiate pulmonary injury accompanying high-grade candidemia.

We tested the hypothesis that regulation of tumor necrosis factor-alpha (TNF-alpha) and IL-6 by the liver differs after intraportal challenge with Candida albicans spp. vs. gram-negative or gram-positive bacteria, independent of microbial clearance kinetics or hepatic O2 consumption (VO2). Buffer-perfused rat livers were infected with equivalent inocula (10(9) colony-forming units) of viable Escherichia coli serotype 055:B5 (EC), exotoxin C-producing Staphylococcus aureus (SA), or two strains of yeast phase C. albicans (CA-1 and CA-2). Microbial clearance and circulating cytokine levels were assessed over 180 min while monitoring VO2 and functional parameters, after which organ-based microbial killing, cell-associated TNF-alpha, and cytokine mRNA levels were determined. Compared with saline controls (normal saline solution; NSS), circulating and cell-associated TNF-alpha and TNF-alpha transcripts minimally increased after CA. In contrast, large increases in perfusate TNF-alpha occurred after EC, peaking at 180 min [135 +/- 32 U/ml (mean + SE)], concomitant with rises in cell-associated cytokine and TNF-alpha transcripts (P < 0.01 vs. NSS). Circulating TNF-alpha also rose after SA but neither cell-associated nor mRNA levels exceeded NSS values. There were no pathogen-specific differences in microbial clearance or VO2. IL-6 gene expression paralleled that for TNF-alpha, but IL-6 bioactivity in perfusates was inhibited by TNF-alpha-dependent and -independent mechanisms. We conclude that hepatic TNF-alpha and IL-6 expression are differentially regulated after taxonomically diverse microbial challenges, with E. coli eliciting the strongest and Candida spp. the weakest stimulatory responses.

We analyzed differences in host regulation of tumor necrosis factor-alpha (TNF-alpha) production and pathophysiological responses in conscious rats after infection with two strains of pathogenic Candida albicans spp. (CA-1 and CA-2) compared with Escherichia coli serotype 055:B5 (EC). The hypothesis was tested that, in contrast to EC, hypotension, organ injury, and mortality after candidemia are not obligatorily dependent on TNF-alpha or TNF-alpha-induced cyclooxygenase pathway metabolites. Dose, viability, and strain-specific dependencies were established after intravenous 10(6) or 10(9) viable CA, as well as heat-killed (HK) or Formalin-inactivated (FI) CA blastospores, compared with live EC at the 24-h LD25 [10(9) colony-forming units (CFU)] and LD100 (10(10) CFU). Shock without endotoxemia developed 4-8 h after 10(9) live CA-1 or CA-2 (LD100 at 24 h) with disseminated yeast-mycelial transformation and increased microvascular permeability in multiple organs but not after HK or FI CA-1. Peak serum TNF-alpha after an LD100 of CA-1 or CA-2 was < 3% of LD25 EC values and was < 1% of peak values during lethal bacteremia. Similar pathogen-specific differences were found in liver- and lung-associated TNF. Production of functionally inactive TNF-alpha during candidemia was excluded by enzyme-linked immunosorbent assay and sodium dodecyl sulfate-polyacrylamide gel electrophoresis with Western blotting. Passive immunization against TNF-alpha 2 h before microbial challenge was not protective against CA but prevented otherwise lethal EC sepsis. Cyclooxygenase inhibition also failed to attenuate candidemic shock. We conclude that the magnitude and kinetics of TNF-alpha production and TNF-alpha-dependent immunophysiological responses are differentially regulated after lethal fungal vs. gram-negative bacterial infection. Thus TNF-alpha is not a pivotal mediator of the acute Candida septic shock syndrome with disseminated candidiasis.